Predicting service times in emergency departments through process analytics: a case study of the radiology unit

This study aims to develop a forecasting system that provides real-time estimates of service times for the radiological unit of an Emergency Department (ED) and warns managers of any potential deterioration in performance. The system, developed using real data from an Italian hospital, incorporates process mining and machine learning techniques to monitor the current state of the ED processes and predict service times. By doing so, this system enables a real-time monitoring of the radiology unit and a dynamic management of the related activities and resources, facilitating the management of the ED overcrowding.